The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

fdth news

1.5-0 (2026-05-25) - Faria, J. C.

Bug fixes

• make.fdt.simple: relative and cumulative frequencies were computed using length(x) as the denominator, which is incorrect when values fall outside the [start, end] range (they are silently dropped by cut()). The denominator is now sum(f), counting only observations that landed in a class interval.
• quantile.fdt: the quantile-position formula ii * n / (length(probs) - 1) assumed equally-spaced probs. It is now probs[ii + 1L] * n, which works correctly for any arbitrary probability vector.
• mfv.fdt_cat: replaced grep(max(y), y) with which(y == max(y)). grep performs regex string matching and is not suitable for numeric equality tests.
• median.fdt: replaced grep(TRUE, ...) with the idiomatic which(...).

Refactoring

• sd.fdt: method body simplified to sqrt(var.fdt(x, ...)), eliminating the duplicated variance calculation that previously mirrored var.fdt verbatim.
• print.fdt.default / summary.fdt.default: shared formatting logic extracted into the private helper .fdt.format.num; both methods now delegate to it, removing the code duplication.
• print.fdt_cat.default / summary.fdt_cat.default: same pattern — shared logic moved into the private helper .fdt.format.cat.
• make.fdt.format.classes: removed unnecessary invisible() wrapper on the return value.
• fdt.data.frame, make.fdt_cat.multiple, plot.fdt.multiple: replaced 1:length(x) / 1:ncol(x) idioms with seq_along() / seq_len() to avoid off-by-one errors when the input has zero length.

NAMESPACE / imports

• Replaced the catch-all exportPattern(".") with an explicit export() list covering only the public API (fdt, fdt_cat, make.fdt, make.fdt_cat, mfv, sd, var). Internal helpers such as make.fdt.simple, make.fdt.multiple, and make.fdt.format.classes are no longer exported.
• Replaced wholesale import(stats, grDevices, xtable) with selective importFrom() calls, keeping import(graphics) because the plot methods use many low-level primitives from that package.

Documentation

• man/fdt.Rd: removed the data slot from the \value section; that slot was documented but has never existed in the actual return object.
• README.md: project layout corrected — removed references to non-existent /data and /demo directories; added /tests entry.
• DESCRIPTION: minimum R version raised from 2.6.0 (released 2007) to 3.5.0.
• .gitignore: expanded with standard entries for R session files, OS artefacts, editor swap files, and additional vignette build intermediates.

Code style

• Adopted consistent formatting across all R source files, man pages, tests, and vignettes: function calls with more than two arguments are now broken at commas with continuation arguments aligned to the first.
• Added spaces around = (named arguments, formals, assignments) and ensured consistent spacing after commas.
• Consolidated short round() and rep() calls (two arguments) onto a single line for readability.
• Created R/fdt-utils.R to group the shared internal helpers .fdt.format.num and .fdt.format.cat, previously defined inside print.fdt.default.R and print.fdt_cat.default.R.

Tests

• Added tests/testthat/test-mean-median-var-sd.R: 5 tests covering mean.fdt, median.fdt, var.fdt, sd.fdt, and the zero-variance edge case.
• Added tests/testthat/test-fdt-cat.R: 9 tests covering fdt_cat (class, column names, frequency sums, sort behaviour, factor input, print, summary).
• Added tests/testthat/test-fdt-data-frame.R: 7 tests covering fdt.data.frame with multi-numeric columns and the by grouping argument, plus mean, print, and summary on fdt.multiple objects.
• Total test count raised from 11 to 41 (0 failures, 0 warnings).

1.3-4 (2026-05-11) - Faria, J. C.

• Added testthat-based tests under tests/testthat/ (including checks for summary(..., format.classes = TRUE, pattern = "%.2f") using normal R strings, avoiding \% quoting issues from .Rd examples).
• quantile.fdt now returns a named numeric vector (for example, 25%, 50%, 75%), with tests and documentation updates for this behavior.
• Improved vignette didactics in vignettes/latex_fdt.Rnw with a quick-start section, a compact xtable::print arguments guide, reproducible categorical examples, and closing takeaways.
• Refactored mfv.default to return all modal values in multimodal cases.
• Refactored mfv.fdt to detect modal classes robustly and return multimodal results.
• Updated mfv.Rd with multimodal examples, including a deterministic non-adjacent modal-class case.
• Revised package documentation English in .Rd files for clarity and consistency.
• Refactored README.md to a modern style, added CRAN badges, and set lifecycle badge to stable.
• Added NEWS.md in Markdown format.
• Improvements were made to the documentation of the package and functions.

1.3-0 (2023-11-17) - Faria, J. C.

• Adjustment of files encoding as required by the CRAN.
• Adjustments and improvements were made to the documentation of the package and functions.

1.2-6 (2021-10-11) - Faria, J. C.

• The “na.omit” function, previously used in several methods of the package, has been removed. The main use was in the functions: “fdt.default”, “make.ft.multiple”, “fdt_cat.default”, “fdt_cat.matrix” and “fdt_cat.data.frame”.
• Added “na.rm” argument to methods associated with default FALSE. If you have NA values that can influence the structure of the table, the user will receive a error message.

1.2-5 (2020-04-17) - Allaman, I. B.

• Correction of wording of argument x in file fdt.Rd. Removed the snippet “or character/factor (‘fdt_cat’)”.
• The functions latex.fdt, latex.fdt_cat, latex.fdt_cat.multiple, print.latex.fdt and summary.latex.fdt was deprecated. Add the new class to the xtable function: xtable.fdt and xtable.fdt.multiple.
• Corrected bug in plot.fdt.default function, types “cdh” and “cdp”. The relative frequencies not was divide by h (amplitude of the class).
• Add news arguments in the plot.fdt.multiple: grouped - if the “by” argument is different from NULL, then the users have the option the to plot factors in a the same graphics; args.legend - a list with the same arguments of the legend function of the graphics package.

1.2-4 (2018-02-17) - Allaman, I. B.

• Only available on GitHub (under test)
• Remove “tmp” object of the fdt.default.R function in the 51 line. That object was desnecessary.

1.2-3 (2015-12-28) - Allaman, I. B.

• Only available on GitHub (under test)
• Added the “round” argument in the functions latex.fdt_cat and latex.fdt_cat.multiple.

1.2-2 (2015-04-13) - Faria, J. C.

• Only available on GitHub (under test)
• The documentation has been improved.

1.2-1 (2015-04-09) - Allaman, I. B.

• Ivan Bezerra Allaman has began to work in the project as co-author.
• Added the “by” argument to fdt_cat.data.frame function. As consequence, the print.fdt_cat.multiple, plot.fdt_cat.multiple, make.fdt_cat.multiple, fdt_cat.matrix functions were changed.
• Was added the latex.fdt_cat.multiple function to give suport the objects of the fdt_cat.multiple class.
• The Rd files make.fdt_cat.simple, make.fdt_cat.multiple, make.fdt.format.classes, make.fdt.multiple and make.fdt.simple were moved from the documentation.

1.2-0 (2014-10-13) - Faria, J. C.

• Thanks to Ivan B. Allaman to the bug fix of the function make.fdt.simple.R.
• It was added basic suport to categorical variables.
• Some bugs were corrected.

1.1-8 (2013-11-22) - Faria, J. C.

• plot.fdth.multiple: x11() was replaced by the portable way dev.new().
• The documentation has been deeply improved.
• Export LaTeX table of the frequency distribution table (latex.fdt).

1.1-7 (2012-04-02) - Faria, J. C.

• Cosmetic changes to meet the requirements of CRAN.

1.1-6 (2011-08-21) - Faria, J. C.

• The S3 methods print.fdt.default and print.fdt.multiple were improved.
• The S3 methods summary.fdt.default and summary.fdt.multiple were improved.

1.1-5 (2011-05-07) - Faria, J. C.

• The S3 methods plot.default and plot.multiple were improved. Both enable adding the count values to all types of plot.

1.1-3 (2011-04-12) - Faria, J. C.

• The MASS package is not essential and was removed from some examples to avoid NOTES in compilation.

1.1-2 (2010-05-27) - Faria, J. C.

• The S3 methods plot.default and plot.multiple were improved.

1.1-1 (2010-05-22) - Faria, J. C.

• The slot data was removed from the fdt object.
• The S3 methods plot.default and plot.multiple were improved.

1.1-0 (2010-05-20) - Faria, J. C.

• Released to CRAN.

1.0-0 (2010-05-16) - Faria, J. C.

• Restrict to testers.

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.